Davide Evangelista is a Junior Assistant Professor at Univeristy of Bologna specializing in computational imaging, machine learning, and numerical optimization, with a particular focus on solving inverse problems through AI-driven approaches. His work lies at the intersection of deep learning, mathematical modeling, and image reconstruction, with applications in medical imaging, cybersecurity, and optimization. He teaches computational imaging and machine learning, providing students with hands-on experience in addressing real-world inverse problems. He also actively contributes to open-source software development, including IPPy, a Python library for inverse problems which allows for authomatically differentiate through linear operators.
Research Interests
His research interests include (but are not limited to):
-
Neural Network Stability and Robustness in Imaging Applications: Davide investigates the theoretical and empirical stability of deep learning models in imaging, focusing on ensuring that neural networks remain robust to perturbations, noise, and adversarial attacks. His research aims to improve the reliability of AI models in critical applications such as medical imaging and remote sensing. 📄 Key paper: To be or not to be stable, that is the question: understanding neural networks for inverse problems
-
Diffusion Models for Generative Image Reconstruction: His work in diffusion models explores how probabilistic approaches can be used to reconstruct high-quality images from noisy or incomplete data. These methods are particularly relevant for tasks such as super-resolution, denoising, and medical image enhancement. 📄 Key paper: To be published soon…
-
Continual Learning for Adaptive AI Models: Davide is interested in developing machine learning systems that can learn continuously from streaming data without suffering from catastrophic forgetting. His work integrates buffer-based methods, such as A-GEM-like approaches, into cybersecurity applications, particularly Intrusion Detection Systems (IDS). 📄 Key paper: To be published soon…
Through this website, he shares research papers, teaching materials, and software projects, fostering collaboration and knowledge exchange within the scientific community.
📩 For inquiries or collaborations, feel free to get in touch.